Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Determinants of Poverty in Developing Countries: Factors that Effect the Human Poverty Index Working Paper Human Development and Capabilities Association Annual Conference Lima, Peru September 10, 2009 Heath Prince, MPAff, PhD Candidate The Heller School for Social Policy and Management Brandeis University Abstract This paper is a preliminary exploration of factors that theory and the development literature suggest should contribute to changes in the United Nation Development Program’s Human Poverty Index (HPI) for developing countries. It is primarily concerned with inteventions that relate to economic growth, the expansion of human capabilities, and the development of assets. The study focuses on the outcomes of a panel data set of HPI scores for a set of 108 developing countries, between 1998 and 2007. Several models are created to empirically test the relationship between the HPI and indicators relating to economic growth-based, capabilities-based and asset-based approaches to development. Preliminary findings suggest that growth-based interventions have a mixed effect on poverty reduction, depending on the level of a country’s deprivation. Education, livestock value, and the employment to population ratio all tend to reduce deprivation, while other variables included in these models suggest no relationship to deprivation. Introduction “In the economic and social realms, the dignity and complete vocation of the human person and the welfare of society as a whole are to be respected and promoted. For man is the source, centre, and purpose of all economic and social life.” (Gaudium et Spes §63) The debate over the ability of national output metrics to accurately identify what is actually occurring on the ground in terms of human development and human well-being has waxed and waned over the decades since the Bretton Woods Agreement established a system of international aid for developing countries in 1944. Critics argue that using aggregate metrics of economic growth, such as Gross Domestic Product, to measure development is akin to using a telescope where a microscope is required—at best, they capture movements in average economic performance, but are ill-equipped to say anything more about the people who are said to benefit, or suffer, from changes in the broader economy. Part and parcel of this debate has been the proliferation of metrics designed to look beyond monetary-based metrics of development and to identify indicators that better explain how people, rather than economies, are developing. Deeply influenced by the growth the human capabilities approach to development, the United Nations Development Program has been at the forefront of 2 the creation of several composite indices designed to better understand, to one degree or another, some of the more human-specific components of development; the UNDP’s Human Development Index is a composite measure of access to knowledge and longevity, in addition to GDP per capita (United Nations Development Program [UNDP], 2008a); and its Gender-related Development Index and Gender Empowerment Index attempt to account for differences between men in women in their ability to benefit from development. But these are only a few among many composite indices that attempt to capture the multidimensional nature of poverty.1 The primary concern of this study is the UNDP’s Human Poverty Index. The HPI was designed to tell a different story than the HDI. The HPI value reflects the proportion of people affected by any one of three key deprivations— adult illiteracy, death before age 40, and a composite measure of the percentage of children underweight for their age and the percentage of a population who lack of access to clean water—providing a comparative, multidimensional measure of the prevalence of human poverty (UNDP, 2008b). Where the HDI measures development, the HPI measures deprivation; where the HDI examines the progress of a community, the HPI examines those left out of this progress. The debate over the efficacy of GDP per capita as a reliable proxy for poverty reduction has recently returned to the top of the agenda at development conferences and to the tables of content of international development journals. Nobel laureates Amartya Sen and Joseph Stiglitz recently chaired the development of a report commissioned by France’s President, Nicolas Sarkozy, to determine what might replace GDP as a measurement of human well-being (Stiglitz, Sen and The “basic needs” approach from the 1960s, the Physical Quality of Life Index developed by Morris D. Morris in the 1970s, and the British Household Panel Survey are only a few of early attempts to measure the multidimensional nature of poverty. 1 3 Fittousi, 2009). Motivating this analysis is the authors’ finding that “over reasonably long periods of time, GDP growth can correlate poorly with changes in the non-income dimensions of well-being.” (Stiglitz et al, 2009) These findings are echoed by others who argue that income-based measures of poverty are often poor proxies for a capabilities-based definition of deprivation and disparity. Subramanian points to China, Costa Rica, Kenya, Peru, the Philippines, and Zimbabwe, “which have displayed greater success in reducing human poverty than income-poverty,” and that “point to the possibilities of enhancing achievements in the space of human functionings by routes different from those centered exclusively on income growth and the percolation of that growth to the poor.” (Subramanian in McGillivray, 2007) While some of the same criticisms leveled at monetary-based measurements of development have been applied to the HDI, the HPI has been relatively less controversial in that, while many could argue that the range of indicators that account for human development is highly subjective, and, so, difficult to capture in a composite index of only a few metrics, few would argue that the metrics comprising the HPI are not clearly objective indicators of human deprivation. Introduced in the UNDP’s 1997 Human Development Report, the HPI has remained essentially true to its original formulation (with one exception, which I describe below), making analysis of the factors that contribute to movements along the index over time possible. Since 1997, HPI scores have improved for some states, and have declined in others. Factors that contribute to these changes are little understood, however (UNDP, 2008b; Arimah, 2004). 4 This study is an exploration of those factors that appear to contribute to changes in the HPI in developing countries, and will be primarily concerned with policies and activities that relate to, or are explicitly meant to encourage, economic growth, increased literacy and improved health, and asset development. The study focuses on the outcomes of a panel data set of HPI scores for a set of 108 developing countries, between 1998 and 2007, and a range of indicators that the literature and theory suggest should have an effect on poverty. Several models are created to empirically test the relationship between the HPI and indicators relating to economic growth-based, capabilities-based and asset-based approaches to development. 5 Background There is a broad and deep literature on the relationship between GDP and various factors assumed to be associated with economic growth (Todaro and Smith, 2003; Perkins, Radelet and Lindauer, 2006) but relatively few studies of the relationship between these factors and human development, and fewer still on the relationship between these factors and human poverty. Throughout much of the 1980s and 1990s, the Washington Consensus approach to development focused the field’s attention on the implementation of specific neoliberal interventions designed to open developing economies up to the world market, make national governments more efficient, and create an economic surplus that would filter its way down to the poor, gradually reducing poverty. Well prior to the current global economic contraction, this approach had already begun to be questioned, shaping as it had IMF and World Bank granting and lending policies throughout the 1990s (Jolly, 2003; Committee on Financial Services, U.S. House of Representatives, 2007). Concurrent with the arrival of the Washington Consensus was the beginning of a formulation of human development that is meant to directly improve the lives of the poor by improving those conditions that most contribute to their poverty—education, health as well as income—as opposed to indirectly improving human development through broader economic growth. In recent years, the multidimensional nature of poverty has gained ground on unidimensional monetary metrics. Poverty is recognized as resulting from the deprivation of basic capabilities, contributing to illiteracy, reduced life expectancy, poor health, and a general lack of access to those factors that contribute to well-being (Sen, 1999). Recognizing poverty as a multidimensional phenomenon is a 6 prerequisite to understanding and measuring its causes, as well as informing policy decisions aimed at reducing it. Implicit in this “human development” approach is the notion that the causality between human development and growth runs in the opposite direction—human development is essential for economic growth, but economic growth is not contingent upon human development. Oil discovery in the Niger Delta, or mineral exploitation in Sierra Leone, has undeniably increased GDP per capita, but has arguably done little to improve human development in these countries, just to cite two popular examples. Indeed, economic growth in the sense that it is meant in the common development parlance is a distant secondary consideration, to paraphrase Sen, to providing individuals with the capabilities to achieve what they have reason to want to achieve (Sen, 1999). The Human Development Index and the Human Poverty Index The United Nations Development Program began sponsoring the Human Development Report in 1990, at which time the Human Development Index was introduced as a method for ranking countries along a specific, if limited, set of human development metrics that reflect the UNDP’s understanding of poverty as a multi-dimensional phenomenon. The HDI is a weighted index that measures literacy, longevity, and standard of living (as measured by GDP per capita at Purchasing Power Parity in US dollars) for each country. In 1997, the HDR added the Human Poverty Index to its list of multi-dimensional metrics, in part as a response to concerns that the HDI did not adequately capture what was happening to the poorest of the poor in both developing and developed countries. 7 “The HPI concentrates on the deprivation in the three essential elements of human life already reflected in the HDI: longevity, knowledge and a decent standard of living. The first deprivation relates to survival: the likeliness of death at a relatively early age and is represented by the probability of not surviving to ages 40 and 60 respectively for the HPI-1 and HPI-2 (a poverty index for developed countries). The second dimension relates to knowledge: being excluded from the world of reading and communication and is measured by the percentage of adults who are illiterate. The third aspect relates to a decent standard of living, in particular, overall economic provisioning. For the HPI-1, it is measured by the unweighted average of the percentage of the population without access to safe water and the percentage of underweight children for their age.” (UNDP, 2008b) Notably absent from this list of factors are any that directly measure income or other monetarybased measures of well-being. This was intentional and done to better understand the nonmonetary dimensions of poverty. As such, the HPI is relatively free of the distribution-related shortcomings of the other highly-aggregated income-based measurements of well-being. Human Poverty Index Formula HPI= [1/3(P1 + P2 + P3)]1/ P1= Probability of not surviving to age 40 P2= Adult illiteracy rate P3= Unweighted average of: population not using an improved water source and children underweight for age =3 A country’s HPI score equals the proportion of its population effected by these deprivations, providing a comparative measure for the prevalence of deprivation. 8 Over the twelve years for which data has been compiled for the HPI, a majority of developing countries have been able to demonstrate improvements along the index, but a significant minority have recorded increasing scores, indicating that an increasing proportion of their populations are falling into extreme deprivation (Mali—52.8 in 1997, 56.4 in 2007; Mozambique—48.5, 50.6, respectively; Central African Republic—40.7, 43.6, respectively, among others). Moreover, a small number of countries have recorded increases in poverty as measured by the HPI at the same time that average income, as measured by GDP/capita has increased, calling into question the notion that increasing average income equates with decreasing poverty. What drives these changes in the index is the motivating question for this study. Note on study design All available observations for a subset of the 108 countries classified by the Human Development Report 2008 (HDR 2008a) as either low- or medium-human development countries will be used in this study. Panel data from the selected countries will be used to estimate the empirical relationship between the HPI and a number of independent variables thought to affect it. Three Research Questions The primary purpose of this study is to better understand the factors that may account for changes in developing countries’ HPI scores, and therefore to better target policy and programmatic interventions. Specifically, this study will address the following questions: To what degree do economic growth-based policies affect levels of deprivation in developing countries, as measured by the Human Poverty Index? 9 To what degree do human capabilities-based policies affect the HPI in developing countries? To what degree does the adoption of asset-development policies by developing countries affect the HPI? These research questions comport with three related theories regarding the relationship between the various approaches to development and poverty-related outcomes. Three Approaches to Development Literature and theory suggest that a wide range of factors may contribute to poverty reduction in the developing world, and developing countries have pursued various strategies designed to grow their economies, reduce poverty, or both. This paper will specify models intended to determine the effect that growth-based, capabilitiesbased, and asset-based policies and practices have had on poverty in developing countries. The models are meant to contribute to a deeper understanding of the sources of deprivation— to what extent are they rooted in weak economies, inadequate investment in human development, or the relative absence of tangible and intangible sources of wealth, or some combination of each of these? For each development approach, a specific set of independent variables has been selected to explain change over time in the HPI scores for developing countries. The selection of variables for this study is intentionally focused on those that have some relationship to state policy decisions, rather than those that are naturally occurring (geography, natural disasters, e.g.) or those that are more individual in nature (family size, wages, e.g.). By selecting variables that are the result of policy decisions, it is the aim of this report to suggest that 10 state interventions and state policies have, perhaps, as important a role to play in reducing deprivation as do market and civil society actors. Three Models Growth-based variables and HPI Neoliberal economic theory would suggest that trade liberalization, labor market flexibility, and exchange rate controls, to name a few, should have resulted in poverty reduction, measured in GDP per capita and compared to specified poverty levels, as the economies of developing countries grew (Asian Development Bank, 2004; Dollar and Kraay, 2000). However, while a developing country may have managed to increase its GDP per capita over a period of time, it does not necessarily follow that poverty, as measured by the HPI, will have declined. The first model will test the relationship between the HPI and a set of growth-based variables. I look specifically at the broad measure of GDP/capita (loggdp) in order to test the relationship between growth in income, however derived, and deprivation, as measured by the HPI. I also consider foreign direct investment (FDI), export value (Exval), and inflation (Inf) on the grounds that openness to trade and prescriptions for keeping inflation low have been central pieces of the toolkit of international financial institutions in recent decades (tests for multicollinearity among these four variables suggest that the parameter estimates can be uniquely computed). H1 = growth-based variables have no effect on deprivation, as measured by the HPI 11 Capabilities-based variables and HPI Human development advocates, however, argue that poverty reduction is as much, if not more, a function of investments in human capability development than it is of economic growth policies. Proponents of capabilities-based development place human beings, instead of economies, at the center of the development enterprise, and would point to investments in, and policies that support, improved healthcare, education, and social inclusion as likely to increase well-being and, therefore, address the multi-dimensional nature of poverty (Fukuda-Parr, 2002; Sen,1999; Kuklys, 2005). The second model will test the relationship between the HPI and a set of variables that could arguably proxy for capabilities-based indicators, including percentage of children of primary school age who are enrolled (primschen), expenditures per primary school student as a percent of GDP (ex/studprim%GDP), primary school completion rate (primcomrate), public health expenditure as a percentage of government expenditures (healthex), and the proportion of births attended by a skilled physician (propbirskphys). I include development aid (aidgni) in this set of variables on the grounds that aid plays a significant role in increasing capabilities, particularly in terms of primary education and infant mortality (Mosley and Suleiman, 2007). I also include a variable measuring the percentage of women in a given country’s parliament (wnppercent), on the grounds that women’s participation in national parliaments may be a proxy for social inclusion. H2 = human development-based variables reduce deprivation, as measured by the HPI 12 Assets-based variables and HPI Moser proposes an “asset vulnerability framework” (Moser, 1998) as a means for understanding deprivation, particularly in developing countries. For proponents of an assets-based approach to poverty reduction, well-being is less a function of what one can do, and more a function of what one possesses, and the value of those possessions (Moser 2007; Barrett et al, 2007). Asset advocates make the case that assets—“the stock of financial, human, natural, or social resources that can be acquired, developed, improved and transferred across generations” (Moser, 2007)— are, themselves, what are most important in reducing poverty. Advocates for an asset-based approach categorize policies designed to reduce poverty into two categories. The first of these is a set of “safety net” policies that are designed to mitigate the risks that may be associated with “poverty-perpetuating survival strategies or that provide protection against loss of key assets to effectively insure vulnerable people, including the presently non-poor, against potentially catastrophic downside risk.” The second category of policies is referred to as ‘cargo net’ policies in the assets literature. These are those policies designed “to help the persistently poor: build up their base of productive assets through education, land reform or other means, so that they can reach a minimum threshold of wealth necessary to selffinance or self-insure in ways that do not replicate their initial poverty; improve the productivity of the assets held by the persistently poor through improved technologies or market access, thereby increasing their capacity to generate investible surpluses and to self-finance and self-insure; or, 13 access the finance (insurance and capital) necessary to protect and invest in assets and thereby to relax the constraints that often drive persistent poverty.” (Barrett et al, 2006) The third model will test the relationship between the HPI and a set of variables that measure the level of asset development within a country. Among the assets Moser identifies, “cargo net” related indicators, such as labor and human capital development, are the most readily identified from among national account-level data. Loss of employment typically leads to returns to poverty, and is increasingly a problem with the spread of the HIV/AIDS pandemic. (Barrett et al, 2006). Variables selected to measure the relationship between the HPI and asset-based policies include the employment to population ratio (employratio), on the grounds that human capital acquired through education should reduce the likelihood that individuals experience severe deprivation. To the extent that ownership of livestock assets is a key wealth indicator in many parts of the developing world, changes in livestock productivity (liveprod) may be related to components of the HPI, particularly those related to longevity and weight. (Barrett et al, 2006; another reference, Hoddinott and Little et al. document?). In addition, to the extent that access to financial information is found to be a key component of asset-based “cargo net” policies, I have included a variable that measures access to credit information (credinfo). H3 = asset-based variables have no effect on deprivation, as measured by the HPI Methodology Conceptual Model 14 Conceptually, the model is intended to demonstrate that the HPI is responsive to specific policies adopted by, or conditions that exist in, individual countries. In particular, the model will demonstrate that: HPI values change over time, and Growth, capabilities, and assets-based policies or interventions contribute to increasing/decreasing a country’s HPI This conceptual model may be presented as an equation HPIit= f(zg, zc, za, t)+Εit in which the HPI for country i at time t is explained as a function of zg, a vector of economic growth based variables, zc, a vector of capabilities-based variables, za, a vector of assets-based variables, and t, representing time. The variable time is key since it will show whether a change in policy leads to a change in HPI, potentially providing the strongest evidence of ‘cause’. Test of inter-item covariance and scale reliability As an initial step in the analysis, and using data from the 2007/2008 HDR (HDR 2008a), I conducted a Cronbach’s Alpha test in Stata 10 on the individual factors that comprise the HPI— probability of not surviving until age 40, adult illiteracy rate, and the unweighted average of: population not using an improved water source and children underweight for age— in order to determine the reliability of the index. The average inter-item covariance was 166.92, and the scale reliability coefficient produced by Stata was 0.88, suggesting that the individual items composing this particular combination are well correlated and reflect some single underlying factor. 15 Panel regression model with fixed effects This paper analyzes the effects of economic growth, human capabilities, and asset-based policies on deprivation, as measured by the Human Poverty Index, focusing on the primary research question: what is the effect of these policies and interventions on the HPI for selected developing countries? To examine this question, a panel regression model is applied to measure changes in the HPI between 1998 and 2007 for a subset of developing countries. In addition, I conducted a Hausman test on the data to determine whether a random or fixed effects model would be the most appropriate. The random effects specification is rejected in favor of the fixed effects model (p=.0001), for the capabilities and asset-based models, suggesting that there are unobserved, country-specific factors that are correlated with the dependent variable. A fixedeffects model controls for these unobserved effects, permitting consistent estimations of the coefficients. Adjustments to Index The dependent variable in this analysis is an index—the Human Poverty Index defined above— and, as such, is subject to the limitations of other deterministic indices. Specifically, the individual components of the HPI are presented as additive, but may, in fact, not be independent of each other, suggesting a multiplicative function, instead. Also, one could argue that the different components merit different weights, and that the data used to calculate the various components are subject to various types of error (see Limitations below). 16 Nonetheless, I argue that the limitations with the HPI are no more problematic than those associated with calculating GDP/capita, and the benefits of a measurement of the multidimensionality of poverty, however imperfect, outweigh the costs in loss of precision. In 2001, the Human Development Report Office of the UNDP changed the formula used to calculate the HPI by dropping an indicator—“access to health service.” This change in the formula made any sort of trend analysis between 1998 and 2007 difficult, if not fatally suspect. For the purposes of this study, I have recalculated the HPI for each of the countries in my subset, between the years 1998 and 2000, using the same formula (described above) for each year between 1998 and 2007. While limitations related to data quality and reliability persist, and are beyond the scope of this study to address, the adjustments made to the HPI values for 1998 and 2000 at least make these figures comparable to values reported in later years. Stratifying the HPI In addition to determining the effect that the selected explanatory variables have on the HPI, I am also interested in whether, and to what extent, this effect varied by level of deprivation. To this end, I created a variable (hipov) that identifies those countries with HPI values higher than the mean value for all countries (.28). In addition, some effort is made to better understand regional variations in the effect that each of the three sets of variables—growth-based, capabilities-based, and asset-based—has on the HPI. Data Limitations 17 A significant limitation to this study has to do with the nature of the data. National social statistics and national accounts data—data that form the core of the Human Poverty Index as well as other aggregate measures—are subject to the quality of the data collection capacities of respective governments, as well as to retroactive recalculation by the international agencies that collect the data. By definition, developing countries’ ability to consistently collect reliable and representative data is, in many cases, lacking. This is particularly true for data related to human development. Most of the traditional methods for collecting national-level data—censuses, sample surveys, the Civil Registration System, administrative records, international data sources, and data collected by international NGOs—each have limitations that are compounded in developing countries (Seeta, 2005; Harkness, 2007). A study of the causes of the types of deprivation measured by the HPI invites misspecification and threats to internal validity. It is difficult, at best, to anticipate (never mind find data for) all factors that could contribute to changes in any one of the factors that comprise the HPI. However, this study will argue, and the literature suggests, that there is a finite and measurable set of indicators that could account for much of the change in a country’s HPI over time, and that identifying these indicators is an essential first step in refining policy interventions designed to reduce poverty. Findings Using a panel data model in Stata 10, and in order to test the hypotheses described above, a random-effects regression model was applied to the set of “growth-based” indicators, and a fixedeffects regression model was applied to both the “capabilities-based” and “asset-based” indicators. In addition, regressions are run on the stratified HP for each set of variables. 18 ̐ Model 1a: Growth-based variables Each of the explanatory variables in Model 1 has a statistically significant effect on HPI, with both the natural log GDP per capita and Foreign Direct Investment variables suggesting that a one unit increase in each would result in a 9% (t=-2.99) and .15% (t=-2.43) decline, respectively, in poverty as measured by the HPI. Also as the growth literature would predict, a one unit growth in inflation is associated with 4.8% (t=5.69) increase in the HPI. Interestingly from the perspective of Washington Consensus prescriptions, however, is the 2.5% (t=3.42) increase in the HPI associated with each one-unit increase in the export value index variable. ̐ Model 1b. Growth-base variables, hipov=1 When the model is restricted to only those countries with HPI values above .28 in any given year, however, the effect of logGDP per capita becomes insignificant (t=0.22), as does the export value coefficient (t=.65). Both the FDI (t=-3.54) and inflation (t=7.00) remain significant, and the magnitude of their effects increase slightly; a one-unit increase in the FDI variable decreases the HPI by .2%, and a one-unit increase in the inflation variable increases the HPI by 5.3% ̐ Model 1c. Growth-based variables, hipov=0 For those countries with HPI scores lower than .28, and therefore relatively somewhat better off than the hipov=1 countries, a one-unit increase in the natural log of GDP per capita produces a 20% (t=-3.74) reduction in the HPI. However, unlike in the prior two models, a one-unit increase in FDI in those countries with lower HPI scores produces a .5% (t=2.40) increase in the HPI, suggesting a slight increase in the level of deprivation experienced in a given country. The export and inflation variables were insignificant in this model. 19 Capabilities-based variables ̐ Model 2a: Capabilities-based variables Regressing the HPI on the set of capabilities-based variables for all countries for which there was data available suggests that for every percentage increase in primary school enrollment (primschen), the HPI is reduced by 24% (t=-2.52). Coefficients for all other variables in this model were statistically insignificant (however, a 3 year lagged variable for public expenditure on primary school students produces a significant [-3.2] 60% reduction in the HPI) ̐ Model 2b: Capabilities-based variables, hipov=1 When the model is restricted to only those countries with HPI scores above .28, the effect of primary school enrollment increases, reducing the HPI by 39% (t=-3.00). All other coefficients remain insignificant. ̐ Model 2c: Capabilities-based variables, hipov=0 However, when only those countries with HPI scores below .28 are considered, none of the coefficients are found to be statistically significant. Asset-based variables ̐ Model 3a: Assets-based variables In the inclusive assets-based model, livestock productivity appears to have a substantial and significant (t=-2.18) effect on the HPI, reducing it by 26%. All other coefficients were insignificant. ̐ Model 3b: Assets-based variables, hipov=1 20 When only those countries with HPI values above .28 are considered, however, none of the variables in the model appear to be statistically significant. ̐ Model 3c: Assets-based variables, hipov=0 For those countries with HPI values below .28, those with relatively lower levels of deprivation, the story changes considerably. In this model, the employment-to-population variable (t=-2.56) and the livestock productivity variable (t=-2.61) both appear to have substantial effects on the HPI, reducing it by 74% and 18%, respectively. Movement between high and low HPI The distribution of the HPI scores for the 108 countries included in this study is clearly bimodal, suggesting two distinct groups of countries within the larger set, with an approximate mean of .28. In order to understand the tendency of countries to move from one group of relatively high HPI status to relatively low HPI status over time, and vice versa, I created the hipov dummy variable to distinguish between the two groups, with hipov=1 indicating those countries with HPI values above .28. Using the xttrans command in Stata 10, it appears that, of those countries that begin in the relatively lower poverty category, hipov=0, 96.1% remain in this category from year to year, and only 3.9% fall into the higher poverty category. However, of those countries that begin in the relatively higher poverty category, hipov=1, 93% remain in this category, with only 7% moving into the relatively lower poverty category. Conclusion 21 “Understanding what drives the observed HPI measure is crucial in order to prioritise public interventions.” (UNDP, 2008b) If one defines poverty as not merely lack of income, but also, or even rather, the lack of capabilities to pursue of life of one’s own choosing, then some sort of multidimensional measurement of poverty is required. Ideally, this metric would include indicators for education, health, and quality of life, would be regularly and consistently compiled, and would lend itself to examination. While imperfect, the Human Poverty Index is one such metric, and the purpose of this study has been to begin to determine which factors contribute to changing its values over time. This study looked at essentially three sets of indicators—economic growth-based, capabilitiesbased, and asset-based—and attempted to quantify the effect that these indicators have on the HPI in both the entire set of 108 developing countries, and on two sets that differ in terms of the level of deprivation as measured by the HPI. In terms of the economic growth based indicators, when all countries are considered as a single group, the outcomes of a random effects panel regression would seem to uphold findings in the development literature. ̐ ̐ ̐ ̐ Increase in loggdp/cap is associated with a decrease in deprivation Increase in FDI is associated with a decrease in deprivation Increase in inflation is associated with an increase in deprivation Increase in export value index is also associated with an increase in deprivation But for the export value index indicator, these are findings we would expect from most models measuring the effect on income-based metrics of poverty, and are generally consistent with neoliberal, Washington Consensus, growth-based prescriptions. 22 However, when the model is restricted to examining only relatively worse off countries, the picture changes somewhat. ̐ ̐ ̐ ̐ Loggdp/cap appears to have no relationship to hpi Export value index indicator is also statistically insignificant FDI continues to be modestly associated with decreases in poverty Inflation continues to be associated with increases in poverty And when the model is restricted to relatively better off countries, we get yet another story. ̐ An increase in loggdp/cap is associated with a decrease in deprivation ̐ But an increase in FDI is now associated with an increase in deprivation As for the set of capabilities-based variables, none of the models were particularly revealing. Only the primary school enrollment variable appears to have any effect on the hpi. Regarding the asset-based variables, the Livestock Production Index and employment-topopulation variables do seem to be associated with a reduction in deprivation, at least when considering the set of relatively less deprived countries. And while it is somewhat encouraging to find that relatively few countries over the 10 year period covered in this study have slipped backwards in terms of non-monetary metrics of poverty, it cannot but be discouraging to see that so few have managed to improve their status over this period. Additional research is undoubtedly needed, but these preliminary findings do seem to suggest that, contrary to what proponents of neoliberal growth strategies suggest, strategies designed primarily to spur economic growth do not necessarily improve the well-being of the poor. 23 24 References ARIAMAH, BEN C.(2004): “Poverty Reduction and Human Development in Africa.” Journal of Human Development, Vol. 5, No.3 (November). Asian Development Bank and Japan International Cooperation Agency (2004): Economic Growth and Poverty Reduction in Bangladesh. BARRETT, CHRISTOPHER B., MICHAEL R. CARTER, and PETER D. LITTLE (2006): “Understanding and Reducing Persistent Poverty in Africa: Introduction to a Special Issue,” Journal of Development Studies, Vol. 42, No. 2, 167–177, (February) JOLLY, RICHARD (2003): ‘‘Human Development and Neoliberalism, Paradigms Compared,’’ in Sakiko Fukuda-Parr and A. K. Shiva Kumar (eds.) Readings in Human Development: Concepts, Measures and Policies for a Development Paradigm. New Delhi: Oxford University Press. DOLLAR, DAVID AND AART KRAAY (2000): “Growth is Good for the Poor”, World Bank Policy Research Working Paper series, No. 2587, Washington, D.C., World Bank. FUKUDA-PARR, SAIKIKO (2003): “The Human Development Paradigm: Operationalizing Sen’s Ideas on Human Capabilities,” Feminist Economics 9(2 – 3), 2003, 301 – 317 HARKNESS, SUSAN (2007): Social and Political Indicators of Human Well-being, in McGillivray, ed., Human Well-Being: Concept and Measurement, Palgrave Macmillan, Basingstoke, UK KUKLYS, W. (2005): Amartya Sen's Capability Approach: Theoretical Insights and Empirical Applications. Berlin: Springer. MAXIM, PAUL S. (1999): Quantitative Research Methods in the Social Sciences. Oxford University Press. Oxford and New York City MCGILLIVRAY, M. (ed.) (2007): Human Well-Being: Concept and Measurement, Palgrave Macmillan, Basingstoke. MOSLEY, PAUL AND ABRAR SULEIMAN Mosley (2007): “Aid, Agriculture and Poverty in Developing Countries, Review of Development Economics,” 11(1), 139–158, 2007 MOSER, CAROLINE O.N. (1998): “Reassessing urban poverty reduction strategies: The asset vulnerability framework,” World Development Vol 26, No 1, pp 1-19 The World Bank, Washington DC MOSER, CAROLINE O.N. (2007): Reducing Global Poverty: The Case for Asset Accumulation. Brookings Institution Press, Washington, D.C. NOORBAKHSH, F. (1998): “The human development index: Some technical issues and Alternative indices,” Journal of International Development, Vol. 10, 589-605 25 PERKINS, DWIGHT H., STEVEN RADELET AND DAVID L. LINDAUER (2006): Economics of Development, Sixth Edition, W.W. Norton and Company, New York and London. PRADHU, K. SEETA (2005): “Social Statistics for Human Development Reports and Millennium Development Goal Reports: Challenges and Constraints.” Journal of Human Development (6) 3 (November) RANIS, G., F. STEWART. and E. SAMMAN (2006): “Human Development: Beyond the Human Development Index.” Journal of Human Development, Vol. 7, No. 3 (November). SEN, A. (1999): Development as Freedom, Oxford University Press. STIGLITZ, JOSEPH; AMARTYA SEN and JEAN-PAUL FITOUSSI (2009): “Report by the Commission on the Measurement of Economic Performance and Progress.” TODARO, MICHAEL P. AND STEPHEN C. SMITH (2003): Economic Development, Eighth Edition, Addison Wesley, Boston. UNDP (2008a): Human Development Report, Oxford University Press, New York. UNDP (2008b): “Human Development Indices: A Statistical Update 2008,” Oxford University Press, New York. U.S. HOUSE OF REPRESENTATIVES, COMMITTEE ON FINANCIAL SERVICES, REP. BARNEY FRANK, CHAIR. (2007): “The Role and Effectiveness of the World Bank in Combating Global Poverty,” Serial No. 110-33, U.S. Government Printing Office, Washington D.C. WOOLDRIDGE, JEFFREY M. (2006): Introductory Econometrics: A Modern Approach, 3rd Edition. Thomson Southwestern, Mason, OH. 26 Appendix Table 1: Variables and definitions Dependent Variable Human Poverty Index for all Least Developed Countries, 1998-2007 Variables of Interest (1998-2007 for all) GDP per capita/PPP (WDI) Export Value Index (2000=100) Inflation, consumer prices (annual %) Foreign Direct Investment as % of GDP (WDI) United Nations Development Program, Human Development Reports 1990-2009, www.undp.org/hdr Economic growth-based variables GDP per capita based on purchasing power parity (PPP). PPP GDP is gross domestic product converted to international dollars using purchasing power parity rates. An international dollar has the same purchasing power over GDP as the U.S. dollar has in the United States. GDP at purchaser's prices is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Data are in current international dollars. Source: World Bank, International Comparison Program database Export values are from UNCTAD's value indexes or from current values of merchandise exports. Source: United Nations Conference on Trade and Development, Handbook of Statistics, and International Monetary, International Financial Statistics Inflation as measured by the consumer price index reflects the annual percentage change in the cost to the average consumer of acquiring a basket of goods and services that may be fixed or changed at specified intervals, such as yearly. The Laspeyres formula is generally used. Source: International Monetary Fund, International Financial Statistics and data files Foreign direct investment are the net inflows of investment to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor. It is the sum of equity capital, reinvestment of earnings, other long-term capital, and short-term capital as shown in the balance of payments. This series shows net inflows in the reporting economy and is divided by GDP. Source: International Monetary Fund, International Financial Statistics and Balance of Payments databases, World Bank, Global Development Finance, and World Bank and OECD GDP estimates. Capabilities-based variables Development Aid as percent of Gross National Income (WDI) Aid includes both official development assistance (ODA) and official aid. Ratios are computed using values in U.S. dollars converted at official exchange rates. Source: Development Assistance Committee of the Organisation for Economic Co-operation and Development, and World Bank GNI estimates. Per pupil expenditure (primary, secondary) (WDI) Public expenditure per student is the public current spending on education divided by the total number of students by level, as a percentage of GDP per capita. Public expenditure (current and capital) includes government spending on educational institutions (both public and private), education administration as well as subsidies for private entities (students/households and other private entities). Source: United Nations Educational, Scientific, and Cultural Organization (UNESCO) Institute for Statistics. Healthcare Public health expenditure consists of recurrent and capital spending expenditure per capita from government (central and local) budgets, external borrowings (% of government and grants (including donations from international agencies and expenditure) (WHO) nongovernmental organizations), and social (or compulsory) health insurance funds. Source: World Health Organization, World Health Report and updates and from the OECD for its member countries, supplemented by World Bank poverty assessments and country and sector studies. Note: The latest updates on these data are accessible in WHO's National Health Accounts (NHA) website (http://www.who.int/nha/en/) Percentage of primary Net enrollment ratio is the ratio of children of official school age school aged children based on the International Standard Classification of Education 1997 enrolled who are enrolled in school to the population of the corresponding official school age. Primary education provides children with basic reading, writing, and mathematics skills along with an elementary understanding of such subjects as history, geography, natural science, social science, art, and music. Source: United Nations Educational, Scientific, and Cultural Organization (UNESCO) Institute for Statistics. Net enrolment ratio in primary education girls (WDI) Public spending on education, total (% of government expenditure) (WDI) Ratio of girls to boys in primary and secondary education is the percentage of girls to boys enrolled at primary and secondary levels in public and private schools. Source: United Nations Educational, Scientific, and Cultural Organization (UNESCO) Institute for Statistics Public expenditure on education consists of current and capital public expenditure on education includes government spending on educational institutions (both public and private), education administration as well as subsidies for private entities (students/households and other privates entities). Source: United Nations Educational, Scientific, and Cultural Organization 2 Seats held by women in national parliament (%) Equity of public resource use Vulnerability to unemployment Unemployment among secondary school completers Unemployment among tertiary school completers Employment to population ratio (ages 18-24) Livestock production (UNESCO) Institute for Statistics. Women in parliaments are the percentage of parliamentary seats in a single or lower chamber held by women. Source: United Nations, Women's Indicators and Statistics database Equity of public resource use assesses the extent to which the pattern of public expenditures and revenue collection affects the poor and is consistent with national poverty reduction priorities. Source: World Bank Group, CPIA database (http://www.worldbank.org/ida). Asset-based variables Vulnerable employment is unpaid family workers and own-account workers as a percentage of total employment. Source: International Labor Organization, Key Indicators of the Labor Market database. Unemployment by level of educational attainment shows the unemployed by level of educational attainment, as a percentage of the unemployed. The levels of educational attainment accord with the International Standard Classification of Education 1997 of the United Nations Educational, Cultural, and Scientific Organization (UNESCO). Source: International Labour Organization, Key Indicators of the Labour Market database Unemployment by level of educational attainment shows the unemployed by level of educational attainment, as a percentage of the unemployed. The levels of educational attainment accord with the International Standard Classification of Education 1997 of the United Nations Educational, Cultural, and Scientific Organization (UNESCO). Source: International Labour Organization, Key Indicators of the Labour Market database Employment to population ratio is the proportion of a country’s population that is employed. Ages 15–24 are generally considered the youth population. Source: International Labour Organization, Key Indicators of the Labour Market database Livestock production index includes meat and milk from all sources, dairy products such as cheese, and eggs, honey, raw silk, wool, and hides and skins. Source: Food and Agriculture Organization, Production Yearbook and data files. Property rights and rule-base governance Property rights and rule-based governance assess the extent to which private economic activity is facilitated by an effective legal system and rule-based governance structure in which property and contract rights are reliably respected and enforced. Credit information Credit depth of information index measures rules affecting the scope, accessibility, and quality of credit information available 3 through public or private credit registries. The index ranges from 0 to 6, with higher values indicating the availability of more credit information, from either a public registry or a private bureau, to facilitate lending decisions. Source: World Bank, Doing Business project (http://www.doingbusiness.org/). Note: Data are as of June 2008. GINI coefficient (WDI) Share of poorest quintile in national consumption (WDI) Adolescent birth rate Covariates Gini index measures the extent to which the distribution of income (or, in some cases, consumption expenditure) among individuals or households within an economy deviates from a perfectly equal distribution. A Lorenz curve plots the cumulative percentages of total income received against the cumulative number of recipients, starting with the poorest individual or household. The Gini index measures the area between the Lorenz curve and a hypothetical line of absolute equality, expressed as a percentage of the maximum area under the line. Thus a Gini index of 0 represents perfect equality, while an index of 100 implies perfect inequality. Source: World Bank staff estimates based on primary household survey data obtained from government statistical agencies and World Bank country departments. Data for high-income economies are from the Luxembourg Income Study database. For more information and methodology, please see PovcalNet (http://iresearch.worldbank.org/PovcalNet/jsp/index.jsp). Percentage share of income or consumption is the share that accrues to subgroups of population indicated by deciles or quintiles. Percentage shares by quintile may not sum to 100 because of rounding. Source: World Bank staff estimates based on primary household survey data obtained from government statistical agencies and World Bank country departments. Data for highincome economies are from the Luxembourg Income Study database. For more information and methodology, please see PovcalNet (http://iresearch.worldbank.org/PovcalNet/jsp/index.jsp). The adolescent birth rate measures the annual number of births to women 15 to 19 years of age per 1,000 women in that age group. It represents the risk of childbearing among adolescent women 15 to 19 years of age. It is also referred to as the age-specific fertility rate for women aged 15-19. (Millennium Development Goals indicator) Data Sources World Health Statistics 2008. Geneva, World Health Organization, 2008 www.who.int/whosis/whostat/2008/en/index.html 4 National health accounts: country information. Geneva, World Health Organization, 2007. www.who.int/nha/country/en/index.html International Data Base (IDB). Washington, DC, US Census Bureau, 2008 www.census.gov/ipc/www/idb World Development Indicators 2007, the International Bank for Reconstruction and Development, The World Bank, www.worldbank.org/wdi Millennium Development Goals Indicators. The official United Nations Site for MDG Indicators. www.mdgs.un.org/unsd/mdg Freedom House 2007. www.freedomhouse.org/ratings/index.htm 5